Color-Image Quality Assessment: From Prediction to Optimization

Preiss, Jens; Fernandes, Felipe; Urban, Philipp

doi:10.1109/tip.2014.2302684

Cited by 83 publications

(60 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The iColor-Image-Difference (iCID) metric is inspired by SSIM, and was proposed by Lissner et al 44 The metric is designed specially to improve the prediction of chromatic distortions such as those created by gamut-mapping algorithms.…”

Section: Full Reference Image Quality Metrics For Gray Scale Imagesmentioning

confidence: 99%

Image Quality Assessment by Comparing CNN Features between Images

Amirshahi¹,

Pedersen²,

Yu³

2016

jist

View full text Add to dashboard Cite

Abstract.Finding an objective image quality metric that matches the subjective quality has always been a challenging task. We propose a new full reference image quality metric based on features extracted from Convolutional Neural Networks (CNNs). Using a pre-trained AlexNet model, we extract feature maps of the test and reference images at multiple layers, and compare their feature similarity at each layer. Such similarity scores are then pooled across layers to obtain an overall quality value. Experimental results on four state-of-the-art databases show that our metric is either on par or outperforms 10 other state-of-the-art metrics, demonstrating that CNN features at multiple levels are superior to handcrafted features used in most image quality metrics in capturing aspects that matter for discriminative perception.

show abstract

Section: Full Reference Image Quality Metrics For Gray Scale Imagesmentioning

confidence: 99%

Image Quality Assessment by Comparing CNN Features between Images

Amirshahi¹,

Pedersen²,

Yu³

2016

jist

View full text Add to dashboard Cite

show abstract

“…16 Their implementation is publicly available: PSNR and SSIM in the video quality measurement tool, 17 iCID in the supplementary material of. 16 We summarize the mechanism of these metrics as follows PSNR: The PSNR between two images A and B is defined by P SN R(A, B) = 20 log 10 max I RM SE(A, B)…”

Section: Data Sets and Evaluation Metricsmentioning

confidence: 99%

Automatic color correction: region-based approach and performance evaluation using full reference metrics

Beucher

Bilodeau

et al. 2015

J. Electron. Imaging

View full text Add to dashboard Cite

Abstract. This article proposes a solution to automatic color correction between two images/videos based on region correspondences. It starts with image segmentation by marker-controlled watershed transformation, which is faster and produces more uniform regions with better adherence to object boundaries than the segmentation in previous color correction approaches. Then, regions between two images are matched using point feature correspondences which are invariant to geometric transformation and illumination change. Finally, the color distorted image is corrected using the color statistics of corresponding regions and the color transfer functions weighted by influence masks. We demonstrate the experimental results using several data sets and evaluate the color correction by different measures of image similarity.

show abstract

“…They are extracted in the form of IDF maps which depict their spatial organization, and which are then averaged so as to produce a single score for each IDF. We refer to the original paper by Preiss et al 5 for further explanations. Although all maps are originally computed on a single scale, we propose to compute the contrast and structure terms on 5 different scales, as suggested in.…”

Section: Multi-scale Icidmentioning

confidence: 99%

“…Note that unlike state-of-the-art metrics whose parameters are trained over a particular image quality database (e.g. 10 ), these ones were selected by means of a visual inspection by three expert observers, so as to minimize the artifacts when the metric is used as an objective function to optimize gamut mapping (see 5 Section III.F).…”

Section: Multi-scale Icidmentioning

confidence: 99%

“…Later, Preiss et al 5 proposed an improved version -referred to as iCID -in order to account for specific artifacts such as chromatic ringing or chromatic edges. Though the purpose of this most recent publication leans towards optimizing gamut mapping rather than IQA, the iCID metric showed great potential on the 2013 version of the renowned Tampere Image Database (TID2013). 2 Compared to the original CID, this improved version uses less parameters, does not require any training for these parameters and takes into account a larger variety of artifacts by considering more image-difference features.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Evaluating the multi-scale iCID metric

2015

Self Cite

View full text Add to dashboard Cite

In this study, we investigate the extent to which an image-difference metric based on structural similarity can correlate with human judgment. We introduce a modified version of the recently published iCID metric and present new results over two large image quality databases. It is particularly noteworthy that the proposed metric yields a correlation of 0.861 with mean opinion scores on the 2013 version of the renowned Tampere Image Database, without dedicated parameter optimization.

show abstract

Color-Image Quality Assessment: From Prediction to Optimization

Cited by 83 publications

References 36 publications

Image Quality Assessment by Comparing CNN Features between Images

Image Quality Assessment by Comparing CNN Features between Images

Automatic color correction: region-based approach and performance evaluation using full reference metrics

Evaluating the multi-scale iCID metric

Contact Info

Product

Resources

About